My impression is that lisp would be the archetype
of a language that is designed around it's syntax,
and the difficulties of that. I don't even care whether Lisp has any other good qualities - I'm
not going to write applications in a language that
is represented in a syntax that is that ugly (and
yes, that is subjective), which demonstrates a
point of their project: Tieing the language to a particular syntax will make it awkward to deal with
for someone, because no syntax or representation
will suit everyone, and there will always be cases
where a different representation is easier to deal
with.
You "diff" by having a representation tool that
compare the DAG's or whatever the internal representation is, or by having a representation tool that outputs a textual description that can
be textually compared, or by having a representation tool that attempts to verify the
functional (as opposed to syntactical) changes to
the program, by traversing the internal representation.
You seem to imply that the representation must be graphical - but that is completely contrary to their stated goal: representation independence.
Your tools shouldn't care about how your other tools represent the application to you.
The point is that you can have multiple different
representations of the same program - the program
is not tied to any one representation. Of course
humans need a representation to deal with. But today we're stuck with text or poor IDE's, instead
of being able to say "huh, can't find that damn bug from looking at the textual version, lets try looking at it like a flowchart, or UML diagram,
or download that new whizbang representation I
saw on Freshmeat the other day", and get good results.
Current programming languages are poor at doing that, because they're built around the language
syntax, and the textual representation is the
canonical form of the language. This makes it incredibly difficult to make good integrated tools, because you have to expect that programmes will reorganize the source code from time
to time, and that will affect your tool and how
it deals with the code.
As I see it, the real advantage of something like
Eidola, is focusing on the program in a machine
representation, not as source code. Preferrably
in a structure, such as some form of DAG, that
is easy to manipulate for programs.
That can spring either way: A mathematician may
want an interface to it that gives him a representation with only mathematical symbols. A C programmer may insist on manipulating text. Or you
may end up with some for of completely visual programming.
By decoupling the human readable representation of the language from the system
internal representation of the language implementation, you make it possible for people
to use a multitude of tools that are extremely
difficult to deal with currently, because they
have to do stupid hacks to get around the use
of text, such as inserting lots of "DO NOT EDIT"
comments.
Your suggestion of "teaching" the system what you
want it to do is certainly one interesting approach
that would be much simpler to experiment with with
a system like Eidola.
They don't need to specify a format for the kernel
representation, only a common interchange format
that all kernel implementations are required to
support.
The issue isn't really avoiding a representation
of the language (that would be practically impossible), but to stress that a human readable
textual representation is irellevant - which opens
the door to experimenting with representations
that are much easier for programs to deal with.
You're confusing the issues. Sure, visual programming hasn't succeeded, yet. That does not
mean it never will happen. Just that either the
technology or we aren't ready for it yet.
Saying that visual programming hasn't succeeded
because programming is not maths is like saying
apples are no good because you don't like oranges.
Yes, there are many programming languages out there
that has ended up being esoteric research toys because they faily to take the programmer into account (and instead just take mathematicians into account), but that is a completely separate issue.
The problem with existing languages is that they're
designed to be efficiently represented in text
for a human to read, not to be efficiently manipulated by automated tools, or visualized
graphically, or transformed or generated by other
processes than a human using a text editor.
I see this is an interesting endeavour because it
explicitly tries to decouple the syntax of the
language from the semantics: If you want to, you
could still write a frontend that would give you
a text representation.
But what would you have gained from that? And what
would you gain from using an existing language
if its syntax was ripped away
You're one insensitive jerk if I've ever seen
one. You may not have seen anorexia when you
were a kid, but that doesn't mean it didn't exist - one of the major reasons why anorexia nervosa
is so difficult to deal with is that it is a psychological illness where the person suffering
is absolutely sure that they are fine, and that
they need to loose weight. When that image is
questioned, their reaction is to try to prevent
it by any means, by hiding their bodies in baggy
clothes, by pretending to eat, or claiming they've
already eaten, etc.
Where your argument is falling apart, though, is
when you blame the lack of self-control. Anorexia
nervosa is about excersizing an extreme amount of
self control. For many, the feeling of accomplishing something, of succeeding, of being
able to control their body, and get where they
believe the have to get in order to look good (and
gain self esteem), is what drives them into it.
It starts with a diet, and ends in an obsession
in self control.
Try telling the family or friend of someone who has died from anorexia nervosa that it's not an illness. There are thousands of them.
As for competition, sure some element of competition can be fine. But not when it gets so
extreme it divides instead of brining together. And
not when it's excluding large groups by being centered around a limited type of activity (typically sports), which not everyone is interested in, or able to perform well in.
Seeking help for troubled friends is one (good)
thing. Turning them over to people that are likely
to hand them over to the police and have them
charged for making terrorist treats for blowing
off steam is a completely different (bad) thing.
If any of those are what a kid observes, then yes,
I agree they should tell. If a treat is very specific, and there appears to be actual intent,
or if you know of an actual crime being or having
been comitted, I belive you have an obligation
to tell (and in many countries it is a criminal
offence not to).
And I don't believe that's the issue here.
The issue is whether a kid should snitch about
statements as vague as those mentioned in the article. Statements that almost anyone has used
the equivalent of at some point - without meaning
it. And Katz is right: The concept of encouraging
organized snitching is school is something that
has so far primarily been used in totalitarian
dictatorships, or countries on the way there.
The Hitlerjugend is perhaps one of the most chilling examples. One of the primary means for
Hitler to get control of Germany was to use kids
to get information about parents that were anti-Nazi, and harass them. They effectively shattered the communist party and the social
democratic party and the workers movement that
way - the only real resistance to his power.
Noone is claiming that snitching in US schools is
that bad, or that anyone is planning to use it for
that purpose at this point. But it's a path down
a slippery slope. A generation of kids that are
taught that snitching is good is going to be a lot
more open to other totalitarian influence than one
that is taught about integrity and compassion, and
personal responsibility.
There's this uncomfortable little issue with
a basic principle of justice: An accused should
have the right to confront his/her accuser.
The reason? Knowing who accused you can be a major
issue in being able to demonstrate your innocence.
If you don't know that your accuser is someone who
holds a grudge against you for some reason, it's
hard to start guessing.
It dramatically reduces the safety of anyone (and
there are many) wrongfully accused, by making it
impossible for them to confront the reliability
of the accuser and the accusers testimony.
(Note: I'm not making any statement about the specific case mentioned in the article, as I have
no idea about the details of the case)
That's one of Katz main, often repeated, points:
School violence has actually been going down for
years (and still is). So does that mean that it's
high profile cases that cause you to be willing
to forfeit some liberties, not a (non-existing)
rise in violence in US schools?
Also, your stab as daycare is silly. Daycare has
been much more widespread than it is in the US
for decades in Europe, including in countries with
a lot less violence than in the US. Neglect
is a problem wherever it happens - it may happen
in daycare centers, it may happen as a result of
divorces, or it may just as well happen because
your happily married parents just doesn't care about you, or just doesn't understand your needs.
If people are actually doing something wrong, yes.
If the person involved had made anything resembling
a serious death threat, as opposed to saying something practically any person alive has said
in some form over their lifetime, I'd wholeheartedly agree that it should be reported.
Even in this case it might have almost been ok, if
it hadn't been for the schools reaction, and for
the fact that the kid snitching in this case ought
to know how schools deals with this these days,
and refrained from reporting it.
If it had been a matter of telling it to a teacher,
and having the teacher sit down and talk to the
kid, to see if there were any signs of serious
intent, then it would be tolerable. I'd still
think that the kid doing the snitching should have
thought twice about it, and considered whether there were any likelyhood at all that the "threat"
really was intended as one, or just loose talk.
Did you ever call the police as a kid when some
other kid told you he'd get his father to beat you up?
Right, that was really meant as a serious threat
to your health of course, so you probably ran straight home and called the cops to get the kid
arrested? No? You didn't? Maybe it's because
kids say things they don't mean all the time, and
they rarely follow through.
The main problem is that a kid in no way is
qualified to determine what is a "potential killer", and that there's no credible proof that
a kid saying they want to kill someone is more likely to do it that someone who doesn't say it.
Schools are asking kids to report people without
any scientifically well founded guidelines that
ensures they narrow the target to people with any
real likelyhood of ever comitting a crime.
The problem with encouraging children to snitch,
is that it creates an environment where kids are
encouraged to keep their mouth shut, because whatever they say may give them problems, or end
them up in jail. That has a chilling effect both
on free speech, and the way they react will also
likely have a chilling effect on troubled kids willingness to talk to their classmates or school
official about their problems, out of fear of being
reported and turned over to the police.
What would you rather have? That a kid with psychological problems dare talk to their friends
or their teacher, and get to talk about - and hopefully get help for - their problems, or that
they keep it to themselves until it gets bad enough
that they end up ruining both their own and other
peoples lives with violence?
Sure, you'll eventually catch someone dangerous
by having kids snitching. Just as you'll eventually
catch someone dangerous by picking up random people on the street. But do we have any evidence
that kids that are snitched on would more likely
have comitted a crime that someone who isn't snitched on?
And are you willing to go down the slippery slope
into a society where school, employers and government will know everything about you because
someone at some point thought something about you
was suspicious? And are you willing to accept a
society where a weird taste in music, clothes,
hobbies or behaviour may end you in jail - if only
shortly - because someone thought you might become
a criminal at some point?
It isn't a far stretch for targetting people over
what's been "playground talk" for as long as there's been schools, to targetting people over
other "features" of their personality.
There is a huge difference between an adult making
a very specific statement about intending to kill
someone at a specific time, and a kid saying something like "I want you all to die", or "I'll kill you if you tell anyone". Almost anyone will
have said something like the latter, and almost
noone goes through with it.
The question is where do you draw the line? And
why do people ignore that school related violence
was on a downturn before these measures started
popping up? And why do people believe that a child
is qualified to see the signals of another kid
being dangerous?
The main danger isn't the kid who go around saying
they're going to kill someone. The danger lie in
kids that are deeply psychologically disturbed.
Sometimes the two intersect, often it doesn't.
Have anyone actually checked whether there is a larger intersection between those who say they'll
kill and those who do it, than between say, those
that play football and those who kill?
No. Because it's so much easier to just assume that
anyone who is "different", or anyone that is careless about what they say, is dangerous.
And another issue is what they'll be asked to report next?
What will the school officials, without any psychological training, or actual verification
procedures in place, decide may be "contributing factors" to violence?
It isn't that far fetched that someone may start
thinking about requesting kids to report religious
beliefs (because they're scared shitless about
kids flirting with satanism), music interest (ohhh,
those scary heavy metal and death metal people),
political affiliations (those dangerous anarchists
and other political extremists, on both sides of
the spectrum are surely likely to be plotting politically motivated killings - kids these days),
etc.
Or maybe they'll ask them to report their classmates parents? Because of course they affect
their kids, and before you know it they've turned
them into killers...
How long before parents starts seeing the parallels
to Hitlerjugend?
And how many see the irony in the name of Pinkertons program, in light of the movie "The Wave", about a class experiment that ends up in
pupils snitching on each other in the worst possible fascist style?
Read the article again. What he said was that they
had looked at the opportunity of doing that, and
decided it would cost them around 300 million USD,
if they tried doing that through aquisitions. The
other alternative would be to spend huge resources on building it from scratch, and that for that reason they decided to spin off their Linux division to let someone willing to take the time,
and resources to go in that direction, so they could concentrate on what they make their money on: Their applications.
Keep in mind that Corel is relatively cash strapped. Before the Microsoft investment, they were close to bankrupcy.
Also keep in mind that Corel "usually" keep a
large part of the shares when they spin off a unit.
In the case of the Netwinder, for instance, they
got 25% of then HCC, now Rebel.com. So they'll likely hold on to parts of the Linux spin off as
well.
That's just plain wrong. Redhat has had several
profitable quarters. Either you're misinformed,
or you're intentially spreading FUD. The reason
they're not anymore is because they're burning
cash to get growth. Think about it: What's the point of raising more money if you're profitable
and doesn't plan to spend more than you're making
for a while to grow your business?
None.
As for whether they'll become profitable, time
will tell. As far as I can remember, they intend
to reach profitability sometime this year.
Large amounts of mismanagement during the past presidency were the real root of the problem, with a lack of foreign policy being one of the main issues.
You're joking right? You do know that one of the
main reasons most of the world hoped Gore would
win the US presidential elections (in most European countries, something like 75% of those polled
hoped Gore would win) was that people
hoping to the last minute that George "I'm not quite sure where Europe is, but I'm sure
it's nice" Bush wouldn't win.
If anything the Clinton administrations strongest
side was it's foreign policy, which it actually
cared a lot about. Some say too much - that it
actually hurt them in the US, because it took time
away from internal matters.
You can say a lot of things about the Clinton administration, and a lot of them bad, but
lack of foreign policy is certainly not among
them.
It's interesting you're calling the LS-120 a
commercial dead end, since many PC manufacturers
are offering them as an alternative to normal
floppy drives on their machines. And this drive
stores more data (32MB on normal floppies, and 240
on LS-120 floppies), yet supports the same media
and formats for backwards compatibility.
Basically, leaving out a floppy drive for a Zip
isn't really an option for most people, since they
will deal with a floppy every now and then (if nothing else, lots of Windows drivers still come
on floppies).
So that means with Zip you'll end up using another
bay.
What I think this needs to succeed though, is an
internal version that supports IDE, that can be
booted from, so that it can be installed as a floppy replacement, not an addition.
You didn't read it, did you? It supports 32 MB
on a standard floppy, which the LS-120
doesn't. The good thing about that is that standard floppies are dirt cheap. It also said it supported
up to 240MB as an alternative (but not on standard
floppies).
Actually, socialism isn't "a system", it's a group
of ideologies spanning from far left (marxist
"scientific" socialism) to far right (fascism and
stalinism borrow heavily from socialist ideas),
with a hole slew of different characteristics -
from the most oppressive, antidemocratic systems
you could imagine, to the social-democratic elected
governments that control most of Europe.
The one common denominator is that ideologies that
draw from socialism emphasize that the state has a
responsibility to ensure at least a minimum of
quality of life for it's population, and that certain basic needs should be secured to some extent by the state.
What separates them are why, and how.
Fascist argumentation for this is that the state
is everything, and a strong state should be the
goal for everyone, and to achieve this, the state
must ensure that all members of society are productive.
One of the common social-democratic arguments is
that "common decency" mandates that one should not
let people suffer when society is wealthy enough to
cover basic needs without adversely affecting it's
other citizens.
The Marxist argument is that socialism is a stage on the way to communism, where
control is transferred from the capitalist upper
classes to a state apparatus controlled by the
working classes, and presupposes a socialist revolution where the workers have taken control.
For an interesting (while certainly biased)
presentation of other widely diverging socialist
ideologies, read the last chapter of the Communist
Manifest (available online), which is a critique of different
socialist ideologies that ranges from reactionary
to hopelessly utopian.
"communist nation" is a contradiction in terms.
I suggest you read up a bit on marxist theory.
The Communist Manifesto by Marx/Engels, and State and the Revolution by Lenin are good starters.
And it doesn't matter whether you support it or not - marxist theory, whether abused beyond
recognition or not, has had such an impact on the
world for the last 150 years that it's worth
knowing what it's about, and making up your own
opinion based on the original sources, as opposed to relying on regurgitated conjecture from the
mainstream media.
Slashdot Mirror
My impression is that lisp would be the archetype of a language that is designed around it's syntax, and the difficulties of that. I don't even care whether Lisp has any other good qualities - I'm not going to write applications in a language that is represented in a syntax that is that ugly (and yes, that is subjective), which demonstrates a point of their project: Tieing the language to a particular syntax will make it awkward to deal with for someone, because no syntax or representation will suit everyone, and there will always be cases where a different representation is easier to deal with.
You seem to imply that the representation must be graphical - but that is completely contrary to their stated goal: representation independence. Your tools shouldn't care about how your other tools represent the application to you.
Current programming languages are poor at doing that, because they're built around the language syntax, and the textual representation is the canonical form of the language. This makes it incredibly difficult to make good integrated tools, because you have to expect that programmes will reorganize the source code from time to time, and that will affect your tool and how it deals with the code.
That can spring either way: A mathematician may want an interface to it that gives him a representation with only mathematical symbols. A C programmer may insist on manipulating text. Or you may end up with some for of completely visual programming.
By decoupling the human readable representation of the language from the system internal representation of the language implementation, you make it possible for people to use a multitude of tools that are extremely difficult to deal with currently, because they have to do stupid hacks to get around the use of text, such as inserting lots of "DO NOT EDIT" comments.
Your suggestion of "teaching" the system what you want it to do is certainly one interesting approach that would be much simpler to experiment with with a system like Eidola.
The issue isn't really avoiding a representation of the language (that would be practically impossible), but to stress that a human readable textual representation is irellevant - which opens the door to experimenting with representations that are much easier for programs to deal with.
Saying that visual programming hasn't succeeded because programming is not maths is like saying apples are no good because you don't like oranges.
Yes, there are many programming languages out there that has ended up being esoteric research toys because they faily to take the programmer into account (and instead just take mathematicians into account), but that is a completely separate issue.
I see this is an interesting endeavour because it explicitly tries to decouple the syntax of the language from the semantics: If you want to, you could still write a frontend that would give you a text representation.
But what would you have gained from that? And what would you gain from using an existing language if its syntax was ripped away
Never read the book, but if it's as chilling as the movie it should be worht checking out...
Where your argument is falling apart, though, is when you blame the lack of self-control. Anorexia nervosa is about excersizing an extreme amount of self control. For many, the feeling of accomplishing something, of succeeding, of being able to control their body, and get where they believe the have to get in order to look good (and gain self esteem), is what drives them into it. It starts with a diet, and ends in an obsession in self control.
Try telling the family or friend of someone who has died from anorexia nervosa that it's not an illness. There are thousands of them.
As for competition, sure some element of competition can be fine. But not when it gets so extreme it divides instead of brining together. And not when it's excluding large groups by being centered around a limited type of activity (typically sports), which not everyone is interested in, or able to perform well in.
Seeking help for troubled friends is one (good) thing. Turning them over to people that are likely to hand them over to the police and have them charged for making terrorist treats for blowing off steam is a completely different (bad) thing.
And I don't believe that's the issue here.
The issue is whether a kid should snitch about statements as vague as those mentioned in the article. Statements that almost anyone has used the equivalent of at some point - without meaning it. And Katz is right: The concept of encouraging organized snitching is school is something that has so far primarily been used in totalitarian dictatorships, or countries on the way there.
The Hitlerjugend is perhaps one of the most chilling examples. One of the primary means for Hitler to get control of Germany was to use kids to get information about parents that were anti-Nazi, and harass them. They effectively shattered the communist party and the social democratic party and the workers movement that way - the only real resistance to his power.
Noone is claiming that snitching in US schools is that bad, or that anyone is planning to use it for that purpose at this point. But it's a path down a slippery slope. A generation of kids that are taught that snitching is good is going to be a lot more open to other totalitarian influence than one that is taught about integrity and compassion, and personal responsibility.
The reason? Knowing who accused you can be a major issue in being able to demonstrate your innocence. If you don't know that your accuser is someone who holds a grudge against you for some reason, it's hard to start guessing.
It dramatically reduces the safety of anyone (and there are many) wrongfully accused, by making it impossible for them to confront the reliability of the accuser and the accusers testimony.
(Note: I'm not making any statement about the specific case mentioned in the article, as I have no idea about the details of the case)
Also, your stab as daycare is silly. Daycare has been much more widespread than it is in the US for decades in Europe, including in countries with a lot less violence than in the US. Neglect is a problem wherever it happens - it may happen in daycare centers, it may happen as a result of divorces, or it may just as well happen because your happily married parents just doesn't care about you, or just doesn't understand your needs.
Even in this case it might have almost been ok, if it hadn't been for the schools reaction, and for the fact that the kid snitching in this case ought to know how schools deals with this these days, and refrained from reporting it.
If it had been a matter of telling it to a teacher, and having the teacher sit down and talk to the kid, to see if there were any signs of serious intent, then it would be tolerable. I'd still think that the kid doing the snitching should have thought twice about it, and considered whether there were any likelyhood at all that the "threat" really was intended as one, or just loose talk.
Did you ever call the police as a kid when some other kid told you he'd get his father to beat you up?
Right, that was really meant as a serious threat to your health of course, so you probably ran straight home and called the cops to get the kid arrested? No? You didn't? Maybe it's because kids say things they don't mean all the time, and they rarely follow through.
Schools are asking kids to report people without any scientifically well founded guidelines that ensures they narrow the target to people with any real likelyhood of ever comitting a crime.
The problem with encouraging children to snitch, is that it creates an environment where kids are encouraged to keep their mouth shut, because whatever they say may give them problems, or end them up in jail. That has a chilling effect both on free speech, and the way they react will also likely have a chilling effect on troubled kids willingness to talk to their classmates or school official about their problems, out of fear of being reported and turned over to the police.
What would you rather have? That a kid with psychological problems dare talk to their friends or their teacher, and get to talk about - and hopefully get help for - their problems, or that they keep it to themselves until it gets bad enough that they end up ruining both their own and other peoples lives with violence?
Sure, you'll eventually catch someone dangerous by having kids snitching. Just as you'll eventually catch someone dangerous by picking up random people on the street. But do we have any evidence that kids that are snitched on would more likely have comitted a crime that someone who isn't snitched on?
And are you willing to go down the slippery slope into a society where school, employers and government will know everything about you because someone at some point thought something about you was suspicious? And are you willing to accept a society where a weird taste in music, clothes, hobbies or behaviour may end you in jail - if only shortly - because someone thought you might become a criminal at some point?
It isn't a far stretch for targetting people over what's been "playground talk" for as long as there's been schools, to targetting people over other "features" of their personality.
The question is where do you draw the line? And why do people ignore that school related violence was on a downturn before these measures started popping up? And why do people believe that a child is qualified to see the signals of another kid being dangerous?
The main danger isn't the kid who go around saying they're going to kill someone. The danger lie in kids that are deeply psychologically disturbed. Sometimes the two intersect, often it doesn't.
Have anyone actually checked whether there is a larger intersection between those who say they'll kill and those who do it, than between say, those that play football and those who kill?
No. Because it's so much easier to just assume that anyone who is "different", or anyone that is careless about what they say, is dangerous.
And another issue is what they'll be asked to report next?
What will the school officials, without any psychological training, or actual verification procedures in place, decide may be "contributing factors" to violence?
It isn't that far fetched that someone may start thinking about requesting kids to report religious beliefs (because they're scared shitless about kids flirting with satanism), music interest (ohhh, those scary heavy metal and death metal people), political affiliations (those dangerous anarchists and other political extremists, on both sides of the spectrum are surely likely to be plotting politically motivated killings - kids these days), etc.
Or maybe they'll ask them to report their classmates parents? Because of course they affect their kids, and before you know it they've turned them into killers...
How long before parents starts seeing the parallels to Hitlerjugend?
And how many see the irony in the name of Pinkertons program, in light of the movie "The Wave", about a class experiment that ends up in pupils snitching on each other in the worst possible fascist style?
This is a very slippery slope.
Keep in mind that Corel is relatively cash strapped. Before the Microsoft investment, they were close to bankrupcy.
Also keep in mind that Corel "usually" keep a large part of the shares when they spin off a unit. In the case of the Netwinder, for instance, they got 25% of then HCC, now Rebel.com. So they'll likely hold on to parts of the Linux spin off as well.
None.
As for whether they'll become profitable, time will tell. As far as I can remember, they intend to reach profitability sometime this year.
You're joking right? You do know that one of the main reasons most of the world hoped Gore would win the US presidential elections (in most European countries, something like 75% of those polled hoped Gore would win) was that people hoping to the last minute that George "I'm not quite sure where Europe is, but I'm sure it's nice" Bush wouldn't win.
If anything the Clinton administrations strongest side was it's foreign policy, which it actually cared a lot about. Some say too much - that it actually hurt them in the US, because it took time away from internal matters.
You can say a lot of things about the Clinton administration, and a lot of them bad, but lack of foreign policy is certainly not among them.
You're probably referring to the Indrema Entertainment System.
Ummm. That's where the "Ironic, I thought", fit in...
Basically, leaving out a floppy drive for a Zip isn't really an option for most people, since they will deal with a floppy every now and then (if nothing else, lots of Windows drivers still come on floppies).
So that means with Zip you'll end up using another bay.
What I think this needs to succeed though, is an internal version that supports IDE, that can be booted from, so that it can be installed as a floppy replacement, not an addition.
You didn't read it, did you? It supports 32 MB on a standard floppy, which the LS-120 doesn't. The good thing about that is that standard floppies are dirt cheap. It also said it supported up to 240MB as an alternative (but not on standard floppies).
The one common denominator is that ideologies that draw from socialism emphasize that the state has a responsibility to ensure at least a minimum of quality of life for it's population, and that certain basic needs should be secured to some extent by the state.
What separates them are why, and how.
Fascist argumentation for this is that the state is everything, and a strong state should be the goal for everyone, and to achieve this, the state must ensure that all members of society are productive.
One of the common social-democratic arguments is that "common decency" mandates that one should not let people suffer when society is wealthy enough to cover basic needs without adversely affecting it's other citizens.
The Marxist argument is that socialism is a stage on the way to communism, where control is transferred from the capitalist upper classes to a state apparatus controlled by the working classes, and presupposes a socialist revolution where the workers have taken control.
For an interesting (while certainly biased) presentation of other widely diverging socialist ideologies, read the last chapter of the Communist Manifest (available online), which is a critique of different socialist ideologies that ranges from reactionary to hopelessly utopian.
And it doesn't matter whether you support it or not - marxist theory, whether abused beyond recognition or not, has had such an impact on the world for the last 150 years that it's worth knowing what it's about, and making up your own opinion based on the original sources, as opposed to relying on regurgitated conjecture from the mainstream media.